Dismiss Notice

Psst... Ready to join TalkBass and start posting, make new friends, sell your gear, and more?  Register your free account in 30 seconds.

Could someone check my math on a custom cab...

Discussion in 'Amps and Cabs [BG]' started by Wes Whitmore, Dec 19, 2003.


  1. Wes Whitmore

    Wes Whitmore Supporting Member

    Mar 10, 2003
    Columbus, OH
    I am getting everything in line to make a single driver cab using a Delta 12LF. I am going for 3.0 ft3 net (16"T x 19"W x 18"D internal) and I found that I can use 2) 4" x 4.95" round ports to tune it to 53 Hz. I accounted for the ports @ .05ft3 each ,speaker baskets @.07 , and minimal bracing.
    If you can't tell, it is modeled from the Avatar CM112 (or as best as I can try to do what he did), but just custom sized to what I want to do with it. People really seem to like the response of the real one. I don't want to reinvent the wheel. Does my math look right?
    Finally, has anyone used the flight case material to finish over top of a birch cab? It shouldn't add much weight with just aluminum and formica, and it seems to hold up well with my PA equipment. I am looking for some travel protection and something that is a little unique.
    Wes
     
  2. Dahhhh....

    Ah....


    What?

    Man, that is some serious stuff. Kind of blows me away. But, I know someone who talks like this. Let me ask him.
     
  3. 44me

    44me

    Jun 17, 2002
    Bedford, NH USA
    I can't comment on the design, but your math is pretty close. Don't be too skimpy with the bracing. Carefully done, it won't add that much weight or steal much volume, but it will improve the sound.
     
  4. Thor

    Thor Moderator Staff Member Gold Supporting Member

    If I was you, I would PM Joris on that one.

    Did you run that through 'perfect box' or something like that?

    The calculation is correct,
    16x19x18 = 3.16 ft3

    less .17 for basket & ports = 2.99 ft3.

    One of the articles I saw by Len Moskowitz noted that pillow stuffing style packing in the box adds enough 'virtual volume' to offset the speaker basket size, though I am not sure how he calculated that.

    Good luck with it.
     
  5. Wes Whitmore

    Wes Whitmore Supporting Member

    Mar 10, 2003
    Columbus, OH
    I used the WinISD beta software. Not the best, but at least I can understand it. I build better than I design...
    Oh, who is Joris?
    Thanks,
    Wes
     
  6. Thor

    Thor Moderator Staff Member Gold Supporting Member

    Run the search button on user ' Joris '
    then check his home page.

    He has a lot of homemade stuff there, very nice!
    Works as an engineer in the Netherlands, and is into Death Metal. Very cool guy.

    t
     
  7. Petebass

    Petebass

    Dec 22, 2002
    QLD Australia
    That's about 85 Litres and is maybe a little bit too small for that speaker, but it should still work. Tuned to 53Hz, the cab produced a hump in the frequencies just before the low end starts to roll off. The cab will be +2dB at 70Hz. If that was what you're after, then great. If you are after a flatter response, then we can make some minor changes to smooth the hump..........

    At the moment, the low end will start to roll off at 46Hz (that your -3db point). Group delay is an acceptable 15ms at 50 Hz and cone excursion also looks OK.

    So yep you math is looking OK so far. However, with 4 ports at 4.95in each in diameter, they'll have to be 21.1" long. That's a problem for a cab that's 18" deep. I would consider a couple of changes to your design.

    1) maybe tune the cab to 45Hz instead. This will smooth out that hump and produce a flatter response. It also lowers your -3dB point to 43Hz. Group delay still peaks at 15ms but is now happening at 36Hz. It will therefore be less noticable because it's happening at a point where the cab is already -7db.

    2) Use 2 ports instead of 4. It will make them shorter. Assuming you want to use the 4.95in diameter ports, at 45Hz, the ports are 13.57in long. At 53Hz, they'll be 8.77in long.

    These are just suggestions, of course.........
     
  8. Wes Whitmore

    Wes Whitmore Supporting Member

    Mar 10, 2003
    Columbus, OH
    Pete,
    The above plan was with 2 4x4.95" ports. Sorry for the confusion. Will those work?

    If the box is too small, it is only because I tried to do some reverse engeneering on the avatar cab. WinISD suggested something around 3.8ft3 if I recall correctly.

    On a side note, I always have a tough time designing these things. Once I think I have the speaker I want (which by the way, I read about the betalite after I made this post! It looks to be a better solution all the way around), I mess around with the port until i decide that I really don't know what the curve is supposed to look like. I would figure smoother is better, but I dont know for sure. Boost is sometimes good. Please give me some guidence if you have the time.
    Wes
     
  9. Petebass

    Petebass

    Dec 22, 2002
    QLD Australia
    Yep that's better. According to WinISD Pro, with those 2 ports the cab is tuned to 50Hz. So you're only 1.5dB up at 70Hz now which is close enough.to flat for most people.

    Speaker design is as much an artform as playing bass is. I've still got a lot to learn about this aspect of speaker design. Some companies deliberately create humps at certain frequencies (Eden, Carvin) and as sales will tell you, there's a lot of people who go for that type of sound. From a design point of view it's more complicated that "sticking a speaker in a box that's too small" because that usually creates all sorts of problems in terms of excursion and placing the hump in the right/wrong spot.

    But it that was the definitive answer, there would be no such thing as Acme or Accugroove or any other company that sell so many cabs based of flat response.

    Avatar does everything wrong in terms of "classic" speaker design theory. The cabs are often tuned high and are, in theory, likely to to fail due to speaker excursion when driven hard. They aren't supposed to be deep enough for bass. And those 6dB attenuators on the horns are so primitive that they're not supposed to keep up with more modern designs, again in theory. But Avatar have many satisfied customers who like the sound and don't blow em up.

    It gets confusing. But the beauty of building your own cabs is that you're in control of all that garbage. That can be a bad thing though, especially if you don't know what you want in a speaker.

    The only way to wrap your brain around it all is to experiment. I've got 5 very different speaker cabs now, some of them I made, some I bought. Because they're all so different I've been able to form my own opinions of what sounds best at the sort of gigs I do. Personally I prefer a frequency hump at 100-300 Hx from my 15's and 4x10 cabs, but I like my 2x10's to have less bottom end and a flatter response in the mids and highs. I find myself cutting everything below 70Hz at most gigs therefore I don't bother with all that "flat to 30Hz" theory. That means I have to be wary of speaker excursion, especially when I'm on the 5th string.

    So I take all that into account whe I build my cabs. The result would probably be a disaster if released commercially but I don't care. They sound the way I want them to. Each cab has it's surprises too. You can design till you're blue in the face but design software can't account for the natural colourations and tones of the speaker itself. Every speaker has it's own voice. Two speakers with identical frequency response graphs can sound very different to the naked ear because thay'll have differences in the voicing and phase response.

    Somewhere along the line you've gotta stop designing and just build the thing. Suck it and see. Treat it like a prototype if you like, but make sure you have fun with it and learn from the experience.

    Oh and we want pics :)
     
  10. Wes Whitmore

    Wes Whitmore Supporting Member

    Mar 10, 2003
    Columbus, OH
    Well, I can always mock up some test boxes and tweak them until I get it close to where I want it. I just hope I am starting with the right speaker. i know you guys are always looking for the magic speaker that hasn't been created yet, which is a little discouraging for me I guess. All I can do is read reviews of commercial cabs, try to get ahold of the same drivers, and copy the cab design as much as possible.
    I dont really want to stray too far away from the delta lf, but should I even consider the deltalite 12? It weighs less, seems to extend lower. It just doesnt have as high of a wattage rating. It looks like it can work well in a 3ft enclosure tuned to 42 Hz, but i think I can get 2 of the 10s in the same size enclosure with better results.
    Thanks for your time,
    Wes
     
  11. Petebass

    Petebass

    Dec 22, 2002
    QLD Australia
    The deltalite 12 doesn't extend lower than the 12lf. I managed to get a similar response curve but the deltalight 12's cab had to be closer to 4cu.feet than 3. At 3 Cu.feet and tuned to 42 Hz, it's -3dB at 53Hz. If it were me, I'd be leaning towars the Lf but only slightly and mostly because of the power rating.

    I like the look of those Deltalights 10's. Remember that I like 2x10 cabs that don't have too much bottoms. at 3cu.feet, than 2x10 cab is -3Db at 60Hz.

    To help you decide, I've printed a graph for you with the frequency response for all 3 cabs. The green line is the Delta 12LF (3cu.feet, 50Hz). The orange line is the Deltalite 12 (3cu.Feet, 42Hz). The Red line is the 2x Deltalite 10 (3cu.feet, 55Hz).
     
  12. Wes Whitmore

    Wes Whitmore Supporting Member

    Mar 10, 2003
    Columbus, OH
    Pete,
    Thank you for taking the time to graph all three of those out. I think I must have been a little confused with bGavins earlier thread when he refers to the tuning at 42 and 41 Hz for the deltalites. I looked at the post again, where he has a snapshot of his graph, and I just noticed that it was at -10dB down. Apparently same box, but just further down the graph. On that one, the 12" deltalite extended lower, but if it is all below true F3, does it really matter...

    Would it be possible for you to take a screenshot of the parameters of each speaker. WinISD has inputs for many more T/S parameters than what I am provided with from the Eminence web page, and I am not getting the same results as you. I want to make sure I am on the same page.

    I also like the 2x10s (I currently have a 2x10 GK RBH which I love). The weight of the box will only be a little heavier with them. I would think I would only be gaining "attack" with these 10s though (acoording to the model). Maybe total volume as well (giving each speaker it's true RMS power). I want to run this cab standalone, but I do use PA and in ears on stage.
    By looking at the graphs you gave me, the deltalite 12 is the smoothest cab on paper, but I have a feeling I want that boost before rolloff that the 12LF has. I wonder if 2 cabs stacked on top of one another will change the freq response? I really only want 1 cab for transportation purposes, but I can adapt.

    THanks for your time. You have been a huge help.
    Wes
     
  13. Tinkering with frequency response is only part of the process. So far, nobody is paying attention to the limited Xmax, therefore limited power, these drivers will accept.

    In SBB4 tuning, the 2512 has a thermal rating of 300 watts, and is ENTIRELY out of Xmax excursion below 150 Hz at 300 watts. Power handling ability continues to dwindle to a paltry 38 watts at 65 Hz. This is good for about 110 dB at 1 meter.

    In QB3 tuning, the lowest power handling is 48 watts at 73 Hz.

    Unless you plan on playing very low volume this driver is useless. Xmax = 0.1141" is the limiting factor. The mechanical damage point isn't much higher at 0.3346".

    If you want a Neo driver, look at the Beyma 115ND/W instead. It will handle the full 400 watts without exceeding Xmax from 32 Hz and higher. The F3 of 70 Hz in SBB4 is fine, and will produce 119 SPL maximum, and without farting or self-destructing.
     
  14. Wes Whitmore

    Wes Whitmore Supporting Member

    Mar 10, 2003
    Columbus, OH
    It makes you wonder why they even build it and call it a musical speaker if you cant build an enclosure that will work in a bass guitar application. i guess you could throw multiples in a cab until you are happy with the volume.

    I guess I will stick with the Delta LF. It has been proven to be a good woofer, and is costs much less.

    Thanks,
    Wes
     
  15. There are many manufacturers out there, other than Eminence. I have 1,235 drivers in my spread sheet as of this writing. The intent is having a database of drivers from which to select for a specific need.

    Neo drivers are still in their infancy. As the technology matures, they will probably become the primary source for magnetic motors. Right now, there isn't a wide selection.

    I wouldn't get too miffed over the wattage thing. The high wattage numbers of most drivers is only an indication of power handling capacity before the voice coil melts into goo. It is almost always far higher than the Xmax will support. An exception is: JBL. Nearly all the JBL drivers have very conservative power ratings, yet almost all of them are within Xmax limits at that rated power. Funny coincidence, eh?

    A Deltalite 2510 produces 110 SPL from 58 watts from 35 Hz and higher. It is only down -7dB at 43 Hz, which makes it a tasty, light weight 1x10 driver for moderate volumes. I figure about 30 pounds total, if made from 5/8" ply.

    You can go even smaller with a 1x10 Legend B102 in SBB4 that accepts 74 watts from 39 Hz and up. It produces 111 SPL in a smaller cabinet, but has less bottom. Total weight is about 2 pounds less than the Deltalite 2510, but frequency response is much higher. Bass response is worse, off -11dB at 43 Hz.
     
  16. Wes Whitmore

    Wes Whitmore Supporting Member

    Mar 10, 2003
    Columbus, OH
    I dont need much volume, relative to what my 2x10s already do. I have no idea what the spl is, but my 1001RB seldom has the master above 12, and the woofer amp volume above 3 (no drive turned up, and EQs are always flat). I am more interested in the low freq extension. It doesn't have to be loud, but it needs to be at least as full as my GK2x10 is. I really don't think it is that low compared to other cabs like SWR, but it is perfect for me, so maybe I don't need that much. I do know it will have the low B just fine, but I never have to turn it up to see how loud it will go. I also don't play anything lower than drop D. The B was just messing around. As long as my cab has close to the same response, then I will be happy. Keeping it light would be even better.
    Thanks,
    Wes
     
  17. Petebass

    Petebass

    Dec 22, 2002
    QLD Australia
    Thanks Bruce, you're right I didn't pay enough attention to excursion. I was hoping you'd jump in and help us out on that one.

    I don't fully understand the T/S parameters that Winisd Pro uses to generate it's excursion graphs, therefore I have trouble trusting them.
    Here's a screenshot of the excursion graphs for the 3 drivers. I've changed the signal strangth to 40 watts. It seems to support what Bruce is saying. However.......

    That's another one of those theory Vs practice phenomena that I struggle to wrap my brain around. Of the 1000+ drivers on your spreadsheet, only a handfull pass the test. In theory anyone who uses anything other than JBL, or a selected other few, is going to destroy their speaker the moment they hit their power rating. In practice there are thousands on bass players using Eminence loaded drivers who don't have this problem. One of these days I'll understand what this all means.
     
  18. Exceeding Xmas is distortion, not destruction.
    Exceeding Xmech is destruction.

    :D

    Eminence drivers have a significant Xmech on most models. I have it cataloged for all drivers that publish this spec. For example, Xmech for Delta models is 10x the Xmax spec. Most players drive these cones into distortion, but not hard enough to bottom them.

    The problem with Windows based tools is you cannot see the formulas used to calculate the results. Spreadsheets are not nearly as glamorous, but their guts are fully open to examination.

    Excursion is dual-limited:
    (1) Thermal power to voice coil
    (2) Displacement

    Displacement is frequency-dependent. The lower you go, the more the cone moves. It is pointless comparing results that exceed Xmax, because it is beyond these programs to calculate non-linear response. Perhaps LEAP will do it, but I don't own it.

    When choosing a driver, the trick is looking at F3, Fb, Vb, Xmax, GD, and plotting the excursion at various power levels. Some manufacturers also publish power compression specs; a nice touch. I also investigate at 400 watts, as this is an arbitrary wattage that many amps can deliver to a given cabinet. If I need more noise, I add multiple 400w cabs and amp power as required.
     
  19. Petebass

    Petebass

    Dec 22, 2002
    QLD Australia
    (slaps head) I knew that.

    I'm still having trouble believing a 300w speaker can be distorting at 40w. And a speaker that is used in at least one commercially available cab that I know of. Could it be that people react to "Cab farts" the same way they react to human farts? ......"Whoops, I hope no-one else heard that. I may have to keep my mouth shut about that one".

    but some of the formulas on your spreadsheet appear to be hidden. For example, Is there a formula to determine "Y" or "N" in the FPX column? Similarly the column for USE appears to be hidden. I can't quite work out how the TYPE colums works.

    What is GD?
     
  20. Everything in my spread sheet is either documented with an Inserted Comment at the top of the column, or calculated with a formula. The lookup tables for the alignments are hidden from view, but accessible.

    Kapton voice coil formers and modern adhesives will accept a tremendous amount of power before they melt. In other words, the voice coil can handle far more power than the suspension. Rating a driver by its maximum thermal power is just another example of marketing bullsh!t to play the "mine is bigger than yours" game.

    FPX = full power xmax. "Y" if within xmax at full power across the USE range. i.e. 32 Hz and up for a 5-string.

    GD = group delay

    USE = arbitrary column for fast sorting. See comment. Same for L-S-D columns.

    TYPE = SBB4, BB4 from hidden lookup tables.

    The arbitrary columns employ a series of rules: loudness, cabinet size, delta between 43~89 Hz, and FB in SBB4. I'm tired of manually entering this data (and getting it wrong), so I'm writing a VBA module that will compute all these arbitrary values when I recalculate the sheet. The VBA code and hidden tables are readable, but you have to know how to use Excel to do so.